Float#ceil / #floor / #round / #truncate return type

In CRuby these four methods choose their return class from the runtime value of the ndigits argument:

call	CRuby returns
1.9.round	2 (Integer)
1.9.round(0)	2 (Integer)
1.9.round(-1)	0 (Integer)
1.234.round(2)	1.23 (Float)

i.e. Integer when ndigits <= 0, Float when ndigits > 0.

The rule spinel uses

The return type matches CRuby's value-based rule whenever ndigits is a literal integer -- the common case:

• no argument -> Integer (x.round, x.ceil, ...)
• literal ndigits <= 0 -> Integer (x.round(0), x.round(-1))
• literal ndigits > 0 -> Float (x.round(2))

For an ndigits <= 0 result the float formula is cast to sp_int, so both the value and #class match CRuby (1234.5.round(-1) is the Integer 1230).

Non-literal ndigits

A non-literal ndigits is classified at runtime: the result is a boxed value that is an Integer when ndigits <= 0 and a Float when ndigits > 0, matching CRuby's class exactly.

• x.round(n) with a variable n types as a poly (boxed) value and emits a runtime branch on the sign of n, so both the value and #class match CRuby (x.round(n) with n == -1 returns the Integer 1230).

The one case still classified statically as Float is a splat whose presence and sign are not knowable at all:

• x.round(*args) / x.round(*[]) can't be classified statically -- whether any argument is present, and its sign, are runtime facts the arity model does not carry.

The implementation lives in infer_call (src/analyze_infer.c) and the Float round/floor/ceil/truncate arm (src/codegen_call_recv.c): a literal-integer ndigits is classified by sign, and a non-literal one emits a runtime Integer-or-Float branch as a boxed value.